*Review* **Prediction Models for Successful Sperm Retrieval in Patients with Non-Obstructive Azoospermia Undergoing Microdissection Testicular Sperm Extraction: Is There Any Room for Further Studies?**

**Ettore Caroppo 1,\* and Giovanni Maria Colpi 2**


**Abstract:** Several prediction models for successful sperm retrieval (SSR) in patients with azoospermia due to spermatogenic dysfunction (also termed non-obstructive azoospermia—NOA) have been developed and published in the past years, however their resulting prediction accuracy has never been strong enough to translate their results in the clinical practice. This notwithstanding, the number of prediction models being proposed in this field is growing. We have reviewed the available evidence and found that, although patients with complete AZFc deletion or a history of cryptorchidism may have better probability of SSR compared to those with idiopathic NOA, no clinical or laboratory marker is able to determine whether a patient with NOA should or should not undergo microdissection testicular sperm extraction (mTESE) to have his testicular sperm retrieved. Further research is warranted to confirm the utility of evaluating the expression of noncoding RNAs in the seminal plasma, to individuate patients with NOA with higher probability of SSR.

**Keywords:** non-obstructive azoospermia; sperm retrieval; male infertility; microTESE; prediction model
